Recent models for long range (>2 nm) transfer of electrons and holes through DNA suggest a mechanism that is neither a single long distance tunneling event nor a mechanism strictly due to hopping, but a mixture of the two. From results reported here we argue that any complete model of electron or hole transfer in DNA should include the effects of reversible proton transfer. Reversible proton transfer (primarily between guanine-cytosine base pairs) influences the ability of DNA to trap free radicals, which in turn affects the migration of holes and electrons. We present the annealing characteristics of electrons and holes trapped in crystalline oligodeoxynucleotides irradiated at 4 K and annealed stepwise to room temperature (RT). The annealing profiles are relatively insensitive to DNA conformation, sequence, or base stacking continuity. The packing of the DNA duplexes is known, and it is readily shown that electron and/or hole transfer must be intermolecular. The distances required for tunneling between separate molecules are found to be comparable to the distances required for tunneling within a DNA duplex. The annealing characteristics of DNA are considerably different than those found in crystals of a-Me-mannoside, 5‘dCMP, and 1-Methylcytosine:5-Fluorouracil. This difference is ascribed to a mechanism wherein reversible proton transfer is a rate-limiting step for electron/hole transfer. Reversible proton transfer is, thereby, a "gate" for electron/hole transfer (via tunneling). Because reversible proton transfer is thermally activated, it is proposed that the energetics of this transfer is a dominant factor in determining the thermal annealing profile of DNA. The competing reactions that govern electron/hole migration created by annealing samples irradiated at 4 K are applicable to electron/hole migration at RT. Evidence for this comes from the observation that, in a number of DNA crystals, the free radical species and radical yields are very similar to crystals irradiated at RT compared to those irradiated at 4 K followed by annealing to RT. The proposed mechanism for electron and hole migration through DNA is one where short transfers (=1 nm) occur by tunneling, and tunneling is gated by reversible proton-transfer.